mod/parse_url.php

   1 <?php
   2 require_once('include/Photo.php');
   3
   4 if(!function_exists('deletenode')) {
   5         function deletenode(&$doc, $node)
   6         {
   7                 $xpath = new DomXPath($doc);
   8                 $list = $xpath->query("//".$node);
   9                 foreach ($list as $child)
  10                         $child->parentNode->removeChild($child);
  11         }
  12 }
  13
  14 function parseurl_getsiteinfo($url) {
  15         $siteinfo = array();
  16
  17         $ch = curl_init();
  18         curl_setopt($ch, CURLOPT_URL, $url);
  19         curl_setopt($ch, CURLOPT_HEADER, 1);
  20         curl_setopt($ch, CURLOPT_NOBODY, 0);
  21         curl_setopt($ch, CURLOPT_TIMEOUT, 3);
  22         curl_setopt($ch, CURLOPT_RETURNTRANSFER, true);
  23         curl_setopt($ch,CURLOPT_USERAGENT,'Opera/9.64(Windows NT 5.1; U; de) Presto/2.1.1');
  24
  25         $header = curl_exec($ch);
  26         curl_close($ch);
  27
  28         if (preg_match('/charset=(.*?)\n/', $header, $matches))
  29                 $charset = trim(array_pop($matches));
  30         else
  31                 $charset = "utf-8";
  32
  33         $pos = strpos($header, "\r\n\r\n");
  34
  35         if ($pos)
  36                 $body = trim(substr($header, $pos));
  37         else
  38                 $body = $header;
  39
  40         $body = mb_convert_encoding($body, "UTF-8", $charset);
  41         $body = mb_convert_encoding($body, 'HTML-ENTITIES', "UTF-8");
  42
  43         $doc = new DOMDocument();
  44         @$doc->loadHTML($body);
  45
  46         deletenode($doc, 'style');
  47         deletenode($doc, 'script');
  48         deletenode($doc, 'option');
  49         deletenode($doc, 'h1');
  50         deletenode($doc, 'h2');
  51         deletenode($doc, 'h3');
  52         deletenode($doc, 'h4');
  53         deletenode($doc, 'h5');
  54         deletenode($doc, 'h6');
  55         deletenode($doc, 'ol');
  56         deletenode($doc, 'ul');
  57
  58         $xpath = new DomXPath($doc);
  59
  60         $list = $xpath->query("head/title");
  61         foreach ($list as $node)
  62                 $siteinfo["title"] =  html_entity_decode($node->nodeValue, ENT_QUOTES, "UTF-8");
  63
  64         $list = $xpath->query("head/meta[@name]");
  65         foreach ($list as $node) {
  66                 $attr = array();
  67                 if ($node->attributes->length)
  68                         foreach ($node->attributes as $attribute)
  69                                 $attr[$attribute->name] = $attribute->value;
  70
  71                 $attr["content"] = html_entity_decode($attr["content"], ENT_QUOTES, "UTF-8");
  72
  73                 switch (strtolower($attr["name"])) {
  74                         case "fulltitle":
  75                                 $siteinfo["title"] = $attr["content"];
  76                                 break;
  77                         case "description":
  78                                 $siteinfo["text"] = $attr["content"];
  79                                 break;
  80                         case "dc.title":
  81                                 $siteinfo["title"] = $attr["content"];
  82                                 break;
  83                         case "dc.description":
  84                                 $siteinfo["text"] = $attr["content"];
  85                                 break;
  86                 }
  87         }
  88
  89         $list = $xpath->query("head/meta[@property]");
  90         foreach ($list as $node) {
  91                 $attr = array();
  92                 if ($node->attributes->length)
  93                         foreach ($node->attributes as $attribute)
  94                                 $attr[$attribute->name] = $attribute->value;
  95
  96                 $attr["content"] = html_entity_decode($attr["content"], ENT_QUOTES, "UTF-8");
  97
  98                 switch (strtolower($attr["property"])) {
  99                         case "og:image":
 100                                 $siteinfo["image"] = $attr["content"];
 101                                 break;
 102                         case "og:title":
 103                                 $siteinfo["title"] = $attr["content"];
 104                                 break;
 105                         case "og:description":
 106                                 $siteinfo["text"] = $attr["content"];
 107                                 break;
 108                 }
 109         }
 110
 111         if ($siteinfo["image"] == "") {
 112                 $list = $xpath->query("//img[@src]");
 113                 foreach ($list as $node) {
 114                         $attr = array();
 115                         if ($node->attributes->length)
 116                                 foreach ($node->attributes as $attribute)
 117                                         $attr[$attribute->name] = $attribute->value;
 118
 119                         // guess mimetype from headers or filename
 120                         $type = guess_image_type($attr["src"],true);
 121
 122                         $i = fetch_url($attr["src"]);
 123                         $ph = new Photo($i, $type);
 124
 125                         if($ph->is_valid() and ($ph->getWidth() > 200) and ($ph->getHeight() > 200)) {
 126                                 if ($siteinfo["image"] == "")
 127                                         $siteinfo["image"] = $attr["src"];
 128
 129                                 if($ph->getWidth() > 300 || $ph->getHeight() > 300) {
 130                                         $ph->scaleImage(300);
 131                                         $siteinfo["images"][] = array("src"=>$attr["src"],
 132                                                                         "width"=>$ph->getWidth(),
 133                                                                         "height"=>$ph->getHeight());
 134                                 } else
 135                                         $siteinfo["images"][] = array("src"=>$attr["src"],
 136                                                                         "width"=>$ph->getWidth(),
 137                                                                         "height"=>$ph->getHeight());
 138                         }
 139                 }
 140         } else {
 141                 // guess mimetype from headers or filename
 142                 $type = guess_image_type($siteinfo["image"],true);
 143
 144                 $i = fetch_url($siteinfo["image"]);
 145                 $ph = new Photo($i, $type);
 146
 147                 if($ph->is_valid())
 148                         $siteinfo["images"][] = array("src"=>$siteinfo["image"],
 149                                                         "width"=>$ph->getWidth(),
 150                                                         "height"=>$ph->getHeight());
 151         }
 152
 153         if ($siteinfo["text"] == "") {
 154                 $text = "";
 155
 156                 $list = $xpath->query("//div[@class='article']");
 157                 foreach ($list as $node)
 158                         $text .= " ".trim($node->nodeValue);
 159
 160                 if ($text == "") {
 161                         $list = $xpath->query("//div[@class='content']");
 162                         foreach ($list as $node)
 163                                 $text .= " ".trim($node->nodeValue);
 164                 }
 165
 166                 // If none text was found then take the paragraph content
 167                 if ($text == "") {
 168                         $list = $xpath->query("//p");
 169                         foreach ($list as $node)
 170                                 $text .= " ".trim($node->nodeValue);
 171                 }
 172
 173                 if ($text != "") {
 174                         $text = trim(str_replace(array("\n", "\r"), array(" ", " "), $text));
 175
 176                         while (strpos($text, "  "))
 177                                 $text = trim(str_replace("  ", " ", $text));
 178
 179                         $siteinfo["text"] = html_entity_decode(substr($text,0,350), ENT_QUOTES, "UTF-8").'...';
 180                 }
 181         }
 182
 183         return($siteinfo);
 184 }
 185
 186 function arr_add_hashes(&$item,$k) {
 187         $item = '#' . $item;
 188 }
 189
 190 function parse_url_content(&$a) {
 191
 192         $text = null;
 193         $str_tags = '';
 194
 195         $textmode = false;
 196         if(local_user() && intval(get_pconfig(local_user(),'system','plaintext')))
 197                 $textmode = true;
 198
 199         //if($textmode)
 200         $br = (($textmode) ? "\n" : '<br />');
 201
 202         if(x($_GET,'binurl'))
 203                 $url = trim(hex2bin($_GET['binurl']));
 204         else
 205                 $url = trim($_GET['url']);
 206
 207         if($_GET['title'])
 208                 $title = strip_tags(trim($_GET['title']));
 209
 210         if($_GET['description'])
 211                 $text = strip_tags(trim($_GET['description']));
 212
 213         if($_GET['tags']) {
 214                 $arr_tags = str_getcsv($_GET['tags']);
 215                 if(count($arr_tags)) {
 216                         array_walk($arr_tags,'arr_add_hashes');
 217                         $str_tags = $br . implode(' ',$arr_tags) . $br;
 218                 }
 219         }
 220
 221         logger('parse_url: ' . $url);
 222
 223         if($textmode)
 224                 $template = $br . '[bookmark=%s]%s[/bookmark]%s' . $br;
 225         else
 226                 $template = "<br /><a class=\"bookmark\" href=\"%s\" >%s</a>%s<br />";
 227
 228         $arr = array('url' => $url, 'text' => '');
 229
 230         call_hooks('parse_link', $arr);
 231
 232         if(strlen($arr['text'])) {
 233                 echo $arr['text'];
 234                 killme();
 235         }
 236
 237
 238         if($url && $title && $text) {
 239
 240                 if($textmode)
 241                         $text = $br . $br . '[quote]' . trim($text) . '[/quote]' . $br;
 242                 else
 243                         $text = '<br /><br /><blockquote>' . trim($text) . '</blockquote><br />';
 244
 245                 $title = str_replace(array("\r","\n"),array('',''),$title);
 246
 247                 $result = sprintf($template,$url,($title) ? $title : $url,$text) . $str_tags;
 248
 249                 logger('parse_url (unparsed): returns: ' . $result);
 250
 251                 echo $result;
 252                 killme();
 253         }
 254
 255         $siteinfo = parseurl_getsiteinfo($url);
 256
 257         if($siteinfo["title"] == "") {
 258                 echo sprintf($template,$url,$url,'') . $str_tags;
 259                 killme();
 260         } else {
 261                 $text = $siteinfo["text"];
 262                 $title = $siteinfo["title"];
 263         }
 264
 265         $image = "";
 266
 267         foreach ($siteinfo["images"] as $imagedata)
 268                 if($textmode)
 269                         $image .= '[img='.$imagedata["width"].'x'.$imagedata["height"].']'.$imagedata["src"].'[/img]';
 270                 else
 271                         $image .= '<img height="'.$imagedata["height"].'" width="'.$imagedata["width"].'" src="'.$imagedata["src"].'" alt="photo" />';
 272
 273 /*      if ($image != "") {
 274                 $i = fetch_url($image);
 275                 if($i) {
 276                         require_once('include/Photo.php');
 277                         // guess mimetype from headers or filename
 278                         $type = guess_image_type($image,true);
 279
 280                         $ph = new Photo($i, $type);
 281                         if($ph->is_valid()) {
 282                                 if($ph->getWidth() > 300 || $ph->getHeight() > 300) {
 283                                         $ph->scaleImage(300);
 284                                         $new_width = $ph->getWidth();
 285                                         $new_height = $ph->getHeight();
 286                                         if($textmode)
 287                                                 $image = $br . $br . '[img=' . $new_width . 'x' . $new_height . ']' . $image . '[/img]';
 288                                         else
 289                                                 $image = '<br /><br /><img height="' . $new_height . '" width="' . $new_width . '" src="' .$image . '" alt="photo" />';
 290                                 } else {
 291                                         if($textmode)
 292                                                 $image = $br.$br.'[img]'.$image.'[/img]';
 293                                         else
 294                                                 $image = '<br /><br /><img src="'.$image.'" alt="photo" />';
 295                                 }
 296                         }
 297                 }
 298         }*/
 299
 300         if(strlen($text)) {
 301                 if($textmode)
 302                         $text = $br.'[quote]'.trim($text).'[/quote]'.$br ;
 303                 else
 304                         $text = '<br /><blockquote>'.trim($text).'</blockquote><br />';
 305         }
 306
 307         if($image) {
 308                 $text = $br.$br.$image.$br.$text;
 309         }
 310         $title = str_replace(array("\r","\n"),array('',''),$title);
 311
 312         $result = sprintf($template,$url,($title) ? $title : $url,$text) . $str_tags;
 313
 314         logger('parse_url: returns: ' . $result);
 315
 316         echo $result;
 317         killme();
 318 }